Great improvements!!

46 views
Skip to first unread message

Jon Choy

unread,
Aug 9, 2011, 3:11:31 PM8/9/11
to qsap...@googlegroups.com

   The improvements are working great so far!!  I was able to test out the directionality of VCCS and ammeter probe. I entered values in properties as exponentials. These worked great. I noticed the plot window does not accept exponentials when entering the starting and ending frequency. I work with designs that will sweep out to 1GHz or so. So, not entering it as an exponential is a bit cumbersome; but it is still useable.  Other than that I haven't even tested it out the plotting window to see if I can generate good Bode Plots.

   I would like to suggest an enhancement to the GUI.  Most circuit designers that I know who work on schematic capture tools use hot keys to place wires, objects, rotate them, etc.  Here is an example from LTSpice (a widely used free SPICE tool). It not only gives you hot keys, it gives you the ability to reprogram them. Even CAD tools like Visio have hot keys. Almost every schematic tool that I've used and I've used a lot of them both commercial and free have hot keys. It makes schematic capture go so much quicker. Using the mouse, right clicking it, and then getting to the right menu takes a lot of time even for a few devices. I would like to get to a point of entering about 20 objects or so in a relatively quick manner. I realize you have some; but it would be great to have a few more for most common actions.
 





 The other suggestion I would like to make is to be able to enter the properties of values with abbreviations which are standard SPICE. These are useful for physical values. Also, is there a way to display the values on the SYMBOL? 

SPICE UNITS

The default units for SPICE are volts, amps, ohms, farads, henries, watts etc. You can specify values in decimal form, 0.0056, or exponential form, 5.6e-3. SPICE also recognizes the following abbreviations (I prefer lower case; but this example has them in upper case):

F

E-15

femto

P

E-12

pico

N

E-9

nano

U

E-6

micro

M

E-3

milli

 

 

 

K

E+3

kilo

MEGE+6mega
GE+9giga
TE+12tera

For clarity you can add letters to the abbreviation as in 1U or 1UFARADS and both are read as the value 1e-6.. SPICE processes the first letter after the number and ignores the rest.




 THANK YOU!!



   Jon


image001.png

Michele Caini

unread,
Aug 13, 2011, 10:11:43 AM8/13/11
to qsap...@googlegroups.com
Hi,
I added your suggestions to the TODO list.
Thanks a lot for the mail... The work will starts as soon as possible.
However at the moment the main problem is the Qwt library. The version 6.x has been released from the previous one, version 5.x and we have to do a lot of work to correct our code because of the changes in the library.

Best Regards,
Michele



2011/8/9 Jon Choy <jon.s...@gmail.com>
image001.png

Jon Choy

unread,
Sep 13, 2011, 1:45:12 PM9/13/11
to qsap...@googlegroups.com, michele.caini
Michele,

    I find this feature in SNAP very useful. Basically, the idea is to apply a generic input and output source and then be able to generate multiple transfer functions like Voltage Gain (they refer to it as Kv), Current Gain (Ki), Input Impedance with output shorted (Zin, short), Input Impedance with output open (Zin, open), Output Impedance with input shorted (Zin, short), etc. This enables a designer to do several key analysis with the same schematic. We are often calculating Gain as well as Input and Output impedances. So, it is a serious time savior for being able to do quite a bit of analysis with the same schematic.Take a look and let me know what you think.


 Jon



image001.png

Michele Caini

unread,
Sep 17, 2011, 7:08:01 AM9/17/11
to qsap...@googlegroups.com
Hi Jon,
sorry if I reply to you now, I was a little bit busy during the week.
This is a good improvement The current version of QSapecNG permits to set only one output component. In fact multiple output components require several operation involving different current and voltage graphs. However, we can offer alternative transfer functions for a circuit until the two graphs are the same.
Unfortunately we cannot easily calculate all transfer functions, because of the nature of the symbolic approach.
 
Please, suggest to me a list of interesting functions, from your point of view (of course, the user point of view).
I will include them as a feature within the next release of QSapecNG.
 
We are working on QSapecNG 2.1 as a porting from Qwt5 to Qwt6, from Boost 1.41 to 1.46, from Qt 4.5 to Qt 4.7.
That's why you have not to look to the next release as a major improvements, I'm sorry.
However, we'll try to include as new features as possible.
Do not hesitate to submit your ideas and requests!!
 
Thanks a lot,
Michele

2011/9/13 Jon Choy <jon.s...@gmail.com>
image001.png

polo0691

unread,
Oct 10, 2011, 10:04:53 AM10/10/11
to QSapecNG
It would be great if there was a way to do symbolic analysis based on
a text file rather than having to enter with the schematic capture
tool. I can envision a point where I might want to do symbolic
analysis on schematics that I have created in a circuit design
schematic capture tool and then generate a netlist. The other caveat
to making this practical is that internal node names are not only
numbers but text and with "_" characters as well. What this does is
prevents someone from having to generate the schematic twice in two
different tools. Anyway, it is just a thought.


Jon
Reply all
Reply to author
Forward
0 new messages